I completely disagree with the opening premise, which is not supported by any data, that PhDs "are in short supply."
I don't know Chand John, but if he really is an expert programmer, and was unable to get a programming job in Silicon Valley in today's economy, something else is going on. Either he is very, very picky, or employers are turning him down for reasons other than his PhD.
Professional success has much more to do with focus and assertiveness and ambition and other personal qualities than with grades, test scores, and degrees accumulated. Learned this the hard way myself.
I worked in a university research lab and had more than one PhD student/RA inform me that MatLab destroyed their actual programming skills in other languages.
The article is based around the assumption that people will in general choose to hire people less qualified than themselves when given the choice, but no evidence is given.
It should be self evident: all firms are hierarchies and all employees are hired by superior rather than inferior rank. The more controversial implied judgement is that man seek to outrank women in marriage, in effect making the spouse an inferior 'vp' to 'ceo' relation. Although, there is plenty of data to support such a preference empirically (ie, partiarchy). <dons flame suit>
I understand your explanation, but I'd still like to see evidence because I haven't felt the same. I've interviewed many dozen's of individuals over the past few years, and when I reflect on those that received offers from me they were practically always those with the best credentials, and I can't think of any of them who doesn't have a better education than I had.
I'm likely the exception rather than the rule - but still - I'd like evidence on the assumption, as GP said.
Anecdotal, but: Once when I was interviewed, I was almost rejected by the interviewer because it seemed like I had been programming for longer than he had been (he was thinking out loud).
When he realized he'd started at a younger age, and had maybe a year on me, he decided it WAS OK, that he did have "more experience."
He's a smart guy, but I've always wondered at that -- wouldn't it be BETTER to hire people smarter than you, rather than the alternative?
Its important to distinguish (a) talent from (b) rank. The latter is best thought of as Rank=f(talent, experience). It never makes sense to hire the wrong rank, or to invert the rank in a junior/senior reporting role. Hiring top talent is really a separate qustion, but its implied that A players with a lead in experience will outdistance A player of lesser experience...as they both acquire experience linearly together.[1]
[1] Hence you are more likely to successfully hire and incorporate A talent only if you have it to begin with. Senior rank B talent struggles when junior experience A talent is aquired. The benefit of increasing experience accrues disporportionately to the higher talent A, and eventually a rank conflict ensues as the Junior A group catch up and surpass the (previously) Senior B's. A messy promotion/re-organiztion situation typically ensues.
Interesting point. We certainly did end up clashing in the years we worked together, as I am not shy in expressing my opinions about what needs to be done.
It came to a head at one point, and the result was a reorg so that he was no longer my direct supervisor; I threatened to walk if the situation continued. The manager I reported to next wasn't particularly higher "rank", by your f(), since I had at least equivalent talent and more experience, but he was comfortable with the relationship(instead of acting like he felt threatened by it), and as such made a much better manager.
'Steve [Jobs] believed that A players hire A players—that is people who are as good as they are. I refined this slightly—my theory is that A players hire people even better than themselves. It’s clear, though, that B players hire C players so they can feel superior to them, and C players hire D players. If you start hiring B players, expect what Steve called “the bozo explosion” to happen in your organization.' — Guy Kawasaki
That's definitely the wooliest part of the article. The claim is that the better employers value MBAs most, which pushes PhDs down the hierarchy.
Whether this would actually push them far enough to be effectively unemployable (for instance, down to the level of helpdesk-type jobs, which I imagine aren't really that keen to employ PhDs) is much more doubtful. I don't know enough about the Chinese marriage market to know how well it fits with 'matching down, but all in all I don't think the parallel drawn with PhD employment is justifiable.
There's definitely a shortage of science research jobs in fields like biology and physics. On the other hand, having exposure to math & empirical thinking makes you ridiculously employable outside your original field. This seems to be overlooked in the current discourse about "too many PhDs". If, for example, you get a PhD in physics from a decent university, and you're willing to adapt to a different domain, then you will with very high probability get a well paying job.
Also, I share the skepticism some other people expressed here about Chand John's unsuccessful job hunt. Maybe he only wants a job that's a direct extension of his research? Or maybe he has a very strong personality that spooks his potential employers? Or maybe he just never learned how to code? In the absence of some compelling and uncommon reason, I can't really imagine how a PhD in Computer Science (from a good school!) could fail to get a job right now.
One would think more young PhDs would start companies. I think the challenge comes down to funding.
I was speaking to a friend who is a VC about this. She suggested that a VC would have a hard time funding a group of unproven PhDs. She suggested a career path where one works for Google, Yahoo, and ships a few products and then get the street creds desired by VC. That isn't unreasonable.
What is a bit discouraging is being put into the same camp as a fresh graduate (out of undergrad). A lot of interviews seem to test for CS 101 or the CS algorithms course. Every PhD I've spoken too is frustrated by this. The game to play is spend a month with the Cormen book while cursing yourself for doing a PhD :-D
If you go with the articles premise that the "picky" A class PhDs are the people getting leftover by traditional employment options, what else are these people to do?
if you have a Ph.D. in science/engineering and want to get a job related to your speciality, then yes, that's the case. also, it's a no-brainer since there are very few jobs (in either industry or academia) where you can be paid to work on your ultra-specialized field of expertise.
but if you want to get a well-paying job period that doesn't involve your dissertation work, then there are far more options. in fact, the best paying jobs out of Ph.D. are those that have nothing to do with your dissertation work (e.g., finance, engineering at top firms such as Google/Facebook/etc., management consulting).
I agree there are far more options if you look outside your small field, but I think it's not uncommon for employers to be skeptical of hiring someone with a PhD. Some of their reasons may be borne from experience and some may be simply due to stereotypes of PhD holders (e.g. too easily bored with "normal" work). My personal experience has been that many prospective employers seem to think that someone with a fresh PhD is both over and under-qualified (which is twice as scary?).
What is strange about the Chinese situation is that top women are not chosen _despite an oversupply of men_. You can't make the same argument about there being an oversupply of positions for PhDs. The article itself says there are "limited academic positions," and who would expect a company to hire a PhD when an MBA would suffice? What PhD would feel satisfied with a job an MBA could handle?
This analogy sort of works because despite what we see in media, men aren't typically the choosers when it comes to relationships, especially when there's an oversupply.
In this case PhDs, like women, aren't choosing to settle either.
I suspect the bigger reason is because Ph.Ds are difficult to accurately assess. When a company wants to hire a candidate, they give preference to someone who has a connection to the company or to someone on the inside. Ph.Ds are extremely unlikely to have any meaningful connections in this regard. This makes vetting them difficult as there is no one to vouch for them.
Location of papers is a decent metric, but more and more it's an indicator of whether the Ph.D comes from an 'in lab' in an 'in field'. Quality of publication would be the best metric, but that is extremely difficult for an outsider to judge.
This leads to a shortage as there is a shortage of people a company can reliably hire at the price they want with a skill set that justifies the price.
There's a lot of truth to this. A PhD is to a considerable extent a product of their project, and no two projects are alike, even for students working in the same lab. In fact, a physics problem isn't necessarily going to be solved using physics techniques. My lab mate solved his problem by developing new theory. I solved mine by overwhelming it with electronics and code.
Well someone forward all these overlooked PhDs my way, because when I start my own company I'll have a good selection for a good price then.
(Edit: With some of the replies, I feel I should clarify I am about halfway through getting my own PhD in chemical engineering. I was actually kind of poking fun at the article's assumptions.)
We currently have 3 PhD's that were hired in our department. No actual experience in the job/field, yet they were hired on based solely on their degree. They are doing work than an entry-level employee would be doing. What is the value for their overpaid salary? Their actual job performance? Or the piece of paper that says they went to school for X number of years?
You can have those overlooked PhD's. They are overlooked for a reason.
You realize every PhD isn't the same? People with PhDs in systems are good at building shit. They might very well suck being data scientists. Similarly, someone with a Machine Learning PhD might suck at fault-tolerant middleware.
P.S. The anti-PhD meme on HN really needs to stop.
Often times, PhDs do a PhD because they are deeply interested in a very specific topic. (like, really really specific)
As such, unless your startup tries to solve a problem in a PhD-level domain, it's unlikely that they'd be excited to work for you.
If you want to build the next Kinect, you might find plenty of fresh PhDs who totally rock computer vision research. But if you're building a fairly generic business/consumer app, it's going to be harder.
> If you want to build the next Kinect, you might find plenty of fresh PhDs who totally rock computer vision research. But if you're building a fairly generic business/consumer app, it's going to be harder.
This. After spending many years becoming a world-class expert in X, it's a difficult proposition to go work in the back-office for some boring consultancy firm or SME.
I don't know what the stats are for the PhD market overall, but one issue might be trying to match up with the "highest ranked" company. Unlike MBAs, PhDs are specialized and not fungible. You can't approach a PhD job search the way a Harvard MBA approaches a banking job search (Where there is a clear hierarchy of desirable employers).
The companies I used to work at, one wireless, one telecom, weren't Google or Facebook, but we had a very high ratio of PhDs because we had very specialized needs. If your thesis topic was networking formation algorithms, you might find that a small company working in the mesh networking niche will find you much more valuable relative to a BS than say Facebook.
Hmnn I was surprised at the lack of consideration for specialization -- if someone has spent time to get a PhD in a part of computer science that just doesn't have real-world profitability, it's pretty clear that it's going to be hard to find a job in the market.
Also, I think the very male-driven culture of China was also not discussed enough -- along with effects of new-found freedom found with less glass ceilings.
But definitely a good read -- I don't think I have come in contact with this "matching-down" dynamic as a possible explanation before
> if someone has spent time to get a PhD in a part of computer science that just doesn't have real-world profitability, it's pretty clear that it's going to be hard to find a job in the market.
If they are only looking for jobs in their specialization, that makes sense, but if the PhDs are looking for any job for which they hold the qualifications, it's not obvious that having spent time devoted to specializing ought to be a penalty. If the choice is between someone with a bachelors in CS, and someone with a bachelor's in CS who also has a a PhD in Finnish Basketweaving, why isn't the second person just as hire-able as the first?
Because time spent getting the PhD is time spent away from the field, and that's a negative factor that may or may not be balanced by the specialization of the PhD.
So, yes, all things being equal, I'd definitely want a new CS grad over the basketweaving expert who hasn't touched a computer in 4 years. I'd also take the dev with solid 5 years recent team lead experience over someone who thinks his 5 years on a private project in school merits a senior programmer position on a development platform he's completely ignorant of.
I'm not saying that's the guy in the article, I have no idea.
I think the reason might be a little bit more complex than just whether they have a PhD or not.
Maybe it would make more sense when you consider the person as being 'overqualified', moreso than less hire-able?
Let's say I'm some web startup that focuses on making the best todo list app that has ever been made -- If I get a guy out of college who has little to no experience (but the experience is on the stack that I am using), and I have a PhD with a some specialization like robotics or DSP (vague but bear with me), it might not seem to make as much sense (even given that they have the same experience on the stack that my company uses) to hire the PhD with the robotics/DSP, as he it seems that he is a molded part (specialization), that does NOT go into the hole I'm trying to put him in, whereas the new grad appears to be more pliable (if not incomplete), and will definitely fit, and I can try and build up to expand to fill the role.
Don't know if that analogy made sense, but essentially I mean to say that someone with a PhD might be just as hirable, but from a hiring manager's perspective, hexagon block in a square hole (as opposed to putting a circle block in that square hole)
For me the poster's point was undermined by my own personal experience of interviewing and discovering the alarming frequency at which people with PhDs will bomb the technical interview, simply because they can't program their way out of a paper bag.
(And yes, this is my own personal experience, therefore it doesn't prove anything. If you thought I was trying to prove something, you didn't come close to understanding my comment.)
To make the argument less compelling and more realistic, you have to make the rankings multi-dimensional and draw large tolerance bands around each person's ranking.
But is the market actually bad for PhD scientists in America?
Dismissing the personal anecdotes of Chand John, the numbers posted show that half of PhD engineers have a job on graduation day, i.e., before most of them even start looking generally, and that was true for other sciences up until the big recession.
That looks like a pretty healthy market to me. If there's still high unemployment after 9 month, that might mean something, but I'm not seeing any hardship in the numbers provided.
Unless you eventually get a job as a researcher, a permanent one, or your family is rich enough that you don't really need to work, yes, getting a PhD in the sciences is a poor monetary investment.
Most of the jobs those fresh PhDs are getting are postdocs. They're temporary jobs that pay about as much as a high school teacher. High school teachers will get tenure eventually. Most postdocs will eventually leave academia.
It's bad everywhere. A professor with tenure will hold that job for 20-30 years. In that time, (s)he will graduate at least 5 (in some fields way more) PhD candidates. If the field of specialization of the professor is booming, (s)he may get replaced by 3 new professors. It is way more likely, though, that there will be 20+ PhDs for every single tenure job. Even in 'hot' fields, universities have little incentive to rapidly increase the number of well-paid professors, if phD students and post-docs are willing to do similar work for less money and risk (if that field becomes less hot, you can dump the post-docs, but not the professors)
They can dump the professors now too... tenure doesn't mean as much as it once did. If you are in the sciences tenured professors are now only guaranteed a minimal salary and an office. If you want to keep any sort of lab going, you need to have external (grant) funding.
tl;dr: It depends on who does the picking and the assumptions they hold.
In both cases, the entity with less power (PhD, male) initiates the selection, and the entity (company, female) with greater power says yea or nay.
A way to work around the problem might be for the lesser entities to not pre-select their targets and instead set up an "open picking" a la recess kickball.
It does fly in the face of the philosophy of "hiring people smarter than you", not sure if an analog in the realm of personal relationships exists.
I don't know Chand John, but if he really is an expert programmer, and was unable to get a programming job in Silicon Valley in today's economy, something else is going on. Either he is very, very picky, or employers are turning him down for reasons other than his PhD.
Professional success has much more to do with focus and assertiveness and ambition and other personal qualities than with grades, test scores, and degrees accumulated. Learned this the hard way myself.