PhD-is it really needed for tech company start up or getting R&D job?

1. Mar 7, 2010

Hi everybody,

I'm a new-comer on this phorum.I got a lot of interesting info while reading here but there are some questions which I didn't find answers for,so I wanna moot them to get your advice,guys.

Lemme say about myself a little bit.I'm not a US citizen and living overseas,currently working as Instrument Engineer in oil&gas industry(I got a five-year Engineering degree that matches something medium between BEng and MEng comparing to US gradation).I don't like my work,'cause it's just a field engineer's work without possibility to use my brains totally.I've always tended to research&development work,I love different calculations both in higher Maths and Physics,but I also love Engineering calculations,design and applications. I'd like to develop my skills in these things to get an interesting research job,and then later to start up my own technology company either in Electronis&Electrical Engineering or in the renewable energy field.That's why I've been thinking and gathering information on getting PhD in US and about what I'll be doing after I get it.

Btw,I'm 27 and married with no kids yet but one day we'll hopefully get them,so even though I love my wife,it's sort of constraint and I've gotta be extra-careful before I finally make a decision to leave my current job and dip my head into preparing for TOEFL,GRE,etc.Uuuf...I strongly desire to change my career path...and achieve the goal.So,there are some questions to be asked:

1)Is it really required to be a PhD holder to get some research&development job in some technology company,federal or national lab,whatever?I mean,is it not sufficient to be just a BEng or MEng?
2)Is it possible to start up my own technology company being an BEng or MEng in Electronics&Electrical Engineering,for instance?

I know,you may say that grad school iself can give me a knowledge about nascent technologies,awareness of science and engineering tendencies along with useful contacts and a reputable status for potential investors.But...may be,is there some way to get into R&D work and later create own tech company leaving out PhD?If not,I'll gladly get this degree.
Any comments will be highly appreciated!

2. Mar 7, 2010

Monte_Carlo

You sound really ambitious. I'm in a somewhat similar situation: earned masters in life science and became disenchanted with career prospects. Now I'm taking computer science courses on undergrad level with intention of enrolling in Comp. Sci. Ph.D. program in mid-near feature (3 - 5 years.) I'm also about your age and refrain from settling down (i.e. family, etc) exactly because of gnawing desire for Ph.D.

I presume you are somewhere in Eastern Europe (Vlad / Vladimir) and I live in U.S. So I got to tell you that Ph.D. is of paramount significance if you want to do anything in R&D. From my limited exposure, almost all full-time positions in academia require Ph.D. and quite a few jobs in industry require this degree. And the number of non-academic jobs requiring Ph.D. is growing. The job market is saturated and getting a Ph.D. is pretty much the only chance of getting into R&D work here. This is not likely to revert. Pursuing doctorate will help you bolster your professional network and underline your ambition. I'm aware that in some parts of industry MBAs (2-year graduate degree) actually supervise scientific personnel (something you want to do) but this is rare and will most likely become obsolete.

So my advice: if you are a man with brains - get Ph.D no matter what.

Monte

3. Mar 7, 2010

Choppy

(1) Although this depends on the specific job, the way that I would look at it is: this is your education. A PhD program will give you more skills that apply to the process of research - defining the project, moving from conceptual idea to experimental design, grant writing, data analysis, etc. So if you want to do some sort of research as a career, a PhD is the way to go.

(2) If you're starting your own company, you don't need anything other than an idea and some startup capital. (Although, you might have trouble convincing investors to give you money if you don't have any academic training at all.)

4. Mar 7, 2010

Monte_Carlo

Yeah, this is exactly was I was trying to say.

5. Mar 7, 2010

twofish-quant

It depends on what you define as research and development. Most applications programmers have masters degrees.

Bill Gates and Michael Dell dropped out of college to start their companies so yes it's possible. The skills that you learn in the Ph.D. program are useful, but they are only one skill you need for starting your own company, and probably not the most useful. The most important skill in starting your own company is the ability to talk to someone and get them to hand to you their money.

6. Mar 7, 2010

twofish-quant

A lot depends on the field. In software engineering, finance, and petroleum engineering, relatively few people have Ph.D.'s.

It's not clear to me that this is the case. You have to make the distinction between jobs for which a Ph.D. is useful and those that *require* a Ph.D.

I think it is. Right now the job market is tight because the economy is still a mess, but things are likely to change in about a year or so.

DO NOT GET A PH.D. SOLELY FOR CAREER REASONS. It's like joining the military. Yes it will help you get a job, but don't make it your sole reason for joining. The problem is that if you get a Ph.D. solely for career reasons, you are not likely to finish it.

No. This is standard and is likely to continue because senior management is largely about finance and politics, and those aren't skills for which a Ph.D. is useful. People with Ph.D.'s do make it up the corporate hierarchy, but the Ph.D. itself doesn't help much, and it can hurt you a lot.

Again, this may be different from industry to industry or country to country, but in the fields that I've been in, there is something of a "glass ceiling" for Ph.D.'s. Some of this is self-imposed because I see senior managers dealing and they spend 100% of their time doing non-geek things, and I'd rather do some geeky things.

ABSOLUTELY NOT. Getting a Ph.D. is an extremely, extremely painful process. It's something akin to being in labor for five years. It's rewarding, but it's not for everyone.

7. Mar 7, 2010

shoehorn

This should be scrawled with a broken bottle on the forehead of everyone who thinks they'd like to obtain a Ph. D.

It's a tough, long road and the number of people who truly enjoy the process are few and far between.

8. Mar 7, 2010

Monte_Carlo

What is the ratio of Ph.D. holding quants to non-Ph.D. holders in finance?

9. Mar 7, 2010

twofish-quant

It really depends on what you define as being a "quant". At least in my group, everyone has a masters degree, and the split between Ph.D and masters (mostly CS and applied math) is about 50/50. Someone with just a masters CS degree wouldn't get hired, but someone with CS masters + several years of work experience (not necessarily in finance) would.

Also, outside of the quant hedge funds, people in top management tend not to have Ph.D.'s, and traders and portfolio managers tend to have masters degrees. But it's really hard to tell what that means exactly. There are a few factors that explain why there aren't that many Ph.D.'s in senior management. One is that there just are more far more MBA's issued than Ph.D.'s The US has about 1000 physics Ph.D.'s a year compared to 120,000 MBA's, and Harvard alone graduates about 900 MBA's each year. The other thing is that physics Ph.D's didn't go into finance in large numbers until the 1990's, and the people in top management got their degrees in the late-1970's and early 1980's. Finally there is the fact that a lots of Ph.D.'s just don't want to and don't have to climb the corporate ladder. If it's a choice between making $300K/year doing geeky stuff and$3M/year attending long boring meetings and preparing powerpoints, the number of physics geeks that want to make the really big money is pretty small.

10. Mar 7, 2010

Monte_Carlo

I'm befuddled. I've read 'Life as Quant' by Derman, 'The Quants' by Patterson and a number of other books on quantitative finance. All the key players have doctorate degrees in physics/math/computer science. My understanding was that to dabble in Monte Carlo methodology, design option-valuating models and the like, one would have to be well versed in research methodology. In your experience, what are the roles delegated to master degree holders as compared to doctorates? What kind of non-finance work experience would make CS masters attractive for quant milieu?

11. Mar 7, 2010

twofish-quant

Derman's book is a great as a memoir, but as a description of how quantitative finance is used in finance today, it's hopelessly outdated. I haven't read Patterson's book, but I heard him talk on NPR, and he seemed generally clueless about what a quant is and what they do. He does talk about one part of the story, but is missing the big picture.

In Derman's time (i.e. the 1990's), people used hammers and chisels to hand craft derivative instruments one by one. Today people create derivatives by the ton, and so you can't "hand craft" them. You need large number of automated processes and machines to create those derivatives, and this involves a lot of CS work.

One other analogy is that at one point in the past, people didn't use microcomputers, so when the first microcomputers came along, it was a big deal to use a microcomputer. Today saying your job involves using some microcomputer in finance tells you nothing about what you really do. You can use that analogy for quantitative finance, since it's now so pervasive that saying you are a "quant" is basically the same thing as saying "you use microcomputers" it tells you nothing about what you do, and there are a dozen different jobs that could be argued are "quantish." Also there are tons of jobs on Wall Street that no one argues are quantish, but involve lots of computer programming.

It's not that much delegation. You have the skills to do A, you convince someone to let you do A, you do A. Also, the type of work experience that will get you hired is any work experience.

One curious thing is that there are quantitative hedge funds that will toss your resume in the trash if you have any Wall Street programming experience, and they hire exclusively programmers that have no financial experience, since they think that Wall Street teaches people bad software engineering habits. The thing about the quant hedge funds that Patterson takes about, is that many of them will consider it an insult if you call them "Wall Street" since they are quite proud of not being Wall Street at all. Wall Street in this context means large investment bank.

12. Mar 7, 2010

twofish-quant

Also Patterson talks about the meltdown in August 2007, which was a minor tremor, compared to the big giant earthquake in September, 2008. Because the quant funds blew themselves up in 8/2007, they weren't seriously involved in the September, 2008 collapse of Lehman. Quant funds still exist but they aren't the cash cows that they were in 2006.

One problem is that because Patterson is trying to write an interesting book, he focuses on what he thinks are the interesting part of the story and then misses the parts that he thinks are boring. For example, it Patterson should up on my desk on Friday and asked to see what I was doing, I'd show him a spreadsheet which I'm using to create a project plan for a coding project, and hear me drone on about how my great victory is that I think I know what the headings of the spreadsheet are. He'd yawn and then leave. The problem is that for every Jim Simmons or Cliff Asness there are probably thousands of people doing more or less what I'm doing. So if you focus on the "demi-Gods" you have a seriously, seriously distorted view of what the street looks like.

One other analogy would be if you were a space alien that came to earth and you wanted to know how earthlings live. So you interview Obama, Putin, and Hu jintao. Now you'd come up with some interesting stories, and if you talk to Obama, you might come up with stories that are more interesting than what a used car salesman in Portland, Oregon would tell you, but as a guide to how people on Earth live, it would be totally inaccurate and useless, and that's why in trying to explain what quants do on Wall Street, Patterson's book is pretty inaccurate and useless.

Derman's book has different problem. It's a pretty good description of what quants did in the late-1990's, but that was ten years ago, and it's a pretty bad description of what quants do now, just like a memoir about what an internet startup looks like in 2000 would be useless for figuring out what an internet startup looks like in 2010.

13. Mar 7, 2010

Monte_Carlo

Thanks for taking time and replying at length; I appreciate meaningful responses like yours. I see we've gone on a lengthy tangent from what Vlad's original post was about. I'm not an effective polemicist and won't argue with you over the domain of your expertise; hence I wholly submit to your opinion. Though I have a lingering feeling that in a highly competitive and professional industry like quantitative finance, ambition for career growth is a tacit requirement for continued employment. I'd imagine that holding Ph.D. or aspiring to obtain one is perceived as a manifestation of that ambition.

14. Mar 8, 2010

TMFKAN64

I think I've benefited a lot in industry by having a Ph.D... while it certainly isn't a requirement for any job I've ever had, it *does* have a way of making people take me a little bit more seriously right off the bat.

Is that worth 7 years of my life? Probably not, but that wasn't the reason I pursued a Ph.D. anyway.

As for startups... there is a big difference between working for a startup, and starting your own. If you want to work for a startup, a BEng or MEng is more than enough education. If you want to start your own, a Ph.D. might help you get taken a bit more seriously, but the most important thing is to have a good idea and a good PLAN to bring it to market. (The name of the game isn't developing new technology... it's SELLING new technology.)

Running a company requires a completely different skill set than that of a typical engineer (or Ph.D.).

15. Mar 13, 2010

Hello guys,

Thanks a lot for your feedbacks.Monte,yes,I'm from Eastern Europe,good guess.

Well,I'm sort of confused,'cause information I got about whether PhD is needed or not for me is quite contradictory.There are arguments both "for" and "against" it...It's very difficult for me to make a clear picture of the real situation with PhD in US,'cause I live overseas.But I'm still thinking of getting PhD as opportunity for future better career and business start up regardless of how much I'm time limited and constrainted by the current job and my family.
Another thing come to my mind is...yeah,I understand that PhD isn't as necessary for start up as great business idea and ability to convince investors but...taking into account that I'm not living in the US and I don't possess as much technical information as,for instance,US PhD students which are on the cutting edge,then I might suggest that besides deeper research knowledge and skills,PhD program could give me another things -the idea itself('cause I think you'll agree that proper ideas rarely come to chance people but normally to them who are working in a proper field)and useful contacts for future investments.

And,of course,we shouldn't forget about enjoying the process of study itself.At least,I think I will.I do more care of financial side - being able to live on miserable scholarship.

Waiting for your comments...show must go on!

P.S.Twofish-quant,I'd like to hear your comments as well,if you don't mind.

16. Mar 13, 2010

Monte_Carlo

This is good. I myself study to retake GRE and apply for Ph.D. at some point. Despite industry careers not formally requiring doctorates, more industry jobs implicitly require it. Think of it like a poker game: if you hold an ace, you generally have a better chance of making the winning hand than if you hold a king.

17. Mar 13, 2010

kote

Research jobs require it, but for anything else I wouldn't consider a PhD an ace. Getting a PhD shows that you are interested in research and technical things. If you are interested in management you could spend the same time you would have spent on a PhD becoming an executive or earning an MBA and getting experience etc. You will have a tough time explaining why you spent (wasted?) time on a PhD for anything other than research or technical roles.

Of course if you want to do research or some types of engineering a PhD could definitely make sense.

18. Mar 13, 2010

IMHO,if you're just a manager with MBA working in some technical company and ruling,for example,PhD employees then,I guess,they can make you shut up your mouth from time to time,because they're more advanced in technologies the company deals with.Is it wrong?

Especially,if we're talking about starting up your own tech company,then the starter needs to be quite qualified in technical things which he's trying to bring to the market.So,I suppose even if one is a good organizer with extra excellent communication skills and intends to hire geeks to do a technical work,even in this case he MUST have a good knowledge of this technology.Especially if he is a creator(inventor,whatever) of one particular technology.Going on...

19. Mar 14, 2010

Sankaku

While this may seem to be a glib statement, it is the truth. Most people who pursue academic excellence (PhD's) do not cultivate the social skills that are required for success in business. Some people do, but the type of person who wants a Physics PhD is often quite different.

A PhD is great for some things, but business street-smarts will win every time when it comes to getting a company off the ground. As twofish says, don't get a PhD for career purposes.

20. Mar 14, 2010

kote

It is part of the technical employee's job to explain his work to a manager. The manager will always have the final ruling and will never just shut up or take technical staff on their word. The onus is on the researcher to explain and justify his work. Even when that is done successfully, a manager still may determine that it is not the best use of time or other resources.

Usually technical managers are promoted from within the technical organization. At some point you will have a CEO or someone else above the research or engineering team, and often times that person will have had no engineering or research experience. When there is a nontechnical manager over technical staff at a lower level, you often end up with the odd situation where the researchers make more money than their boss. Being a manager is not necessarily more prestigious or higher paying than being an individual contributor - it's a different set of skills and responsibilities.