Oh hell yes. My wife, 43, with a PhD in theoretical physics, is working as an adjunct. She might have a line on a 1-year visiting professorship next year. They have yet to decide.

Adjuncts earn less per hour than a Wal-Mart cashier, in case you were wondering.

Shortage, my ass. There's no shortage of trained scientists in America. There's just a shortage of anybody who wants to pay them.
posted by Michael Roberts at 12:37 PM on June 14, 2010 [18 favorites]

An odd set of sentences in the article:
"A prime symptom noted by all: a growing aversion of America¡¯s top students ¡ª especially the native-born white males who once formed the backbone of the nation¡¯s research and technical community ¡ª to enter scientific careers. Increasingly, foreign-born technical and scientific personnel on temporary visas staff America¡¯s university labs and high-tech industries."
posted by Houstonian at 12:38 PM on June 14, 2010

There are many parallels between this article and Frank Donoghue's The Last Professors: The Corporate University and the Fate of the Humanities.
posted by Prospero at 12:47 PM on June 14, 2010 [2 favorites]

There's a lot of truth in here, or at least a lot of stuff that resonates with me. IANAPhD, but I'm surrounded by them at work (life sciences, cancer research), and I see the frustration that most of them can expect to experience for their entire careers.

And despite having a third of the post-college education that most of them have, I make a whole lot more money than the eterna-postdocs, because I have software skills. Even though I've got the long end of that stick, I realize that it's immensely unfair and doesn't do science any favors.

So this article does that well. But it's awfully short on solutions. "Stop treating PhDs like slave laborers". Hell yes, but how?
posted by gurple at 12:52 PM on June 14, 2010 [1 favorite]

Interesting. This is the way it's been for a while in the Humanities, I wasn't aware things were in a similar state on the science side - always had a vague sense that jobs and funding were easier to come by due to potential corporate or military research opportunities.
posted by AdamCSnider at 12:53 PM on June 14, 2010

I've been calling bullshit on Gates' claims for years. It's all a cover for companies like Microsoft to hire more H1-Bs at significantly lower wages.
posted by tommasz at 12:53 PM on June 14, 2010 [7 favorites]

"I just want to say one word to you. One word: plasticsbioinfomatics."

Seriously, I have much to add on this topic with what Mrs. slogger and other friends are going through. But first I'd like to RTFA, and I'm running to a meeting at the moment.
posted by slogger at 12:59 PM on June 14, 2010

The problem with research science, especially in the life sciences, is the large amount of manpower required to do good science coupled with the apprenticeship culture where senior scientists expect graduate students to work (and work hard) for little or no money in return for training, which is theoretically supposed to help them become senior scientists themselves. A generation or two ago, this worked perfectly well--bright students out of undergrad spent no more than five years training under more experienced scientists, got a PhD, and then started their own labs. But now the pyramid scheme has reached the stage where the system is no longer sustainable. There are far too many well-trained scientists and not enough positions. Students spend 6 or more years getting paid as little as $25,000 a year to get a PhD which allows them to join the postdoctoral fellow holding pattern at $35,000 a year for 2, 4, 6, or more years until something potentially opens up in a school so small you've never even heard of. When young, bright students see that smart, hard-working people spend nearly nine years of post-secondary education to make $35,000 a year, it's not surprising that they're fleeing science in droves. The solution is to increase the pay and decrease the number of students admitted, but this brings us back to the first problem, where good science requires lots of cheap, and dedicated workers. I think in time, the only people willing to put up with this will be foreign students. But guess where they'll go once they finish their training at North America's best institutions?
posted by reformedjerk at 1:00 PM on June 14, 2010 [12 favorites]

There's a general higher-education bubble. This is part of it.
posted by Chocolate Pickle at 1:03 PM on June 14, 2010 [4 favorites]

Yes. Yes. Yes. Yes.

I found myself yesterday advising an incredibly promising and talented undergrad to not worry about a PhD and just get a masters, as she'll be waaaaaaay more employable more quickly. If, after the masters, she wanted to go on, do it. But don't start that way unless she was completely certain she knew what she was getting into.

But she already knew this. She already didn't want to go through what she saw so many of her mentors doing. She wanted to pursue science, but have a life, and have a solid prospect for her future now, rather than something shaky in 10 years. She had no PhD plans, and didn't expect that to change.

She left my office, and I was really rather sad. In no small part because of knowing that I still have no idea what's going to be for me in two years, despite my degree, my publications, my decent reputation, and my skills. I have hope, and lots of encouragement, but those often fail the 'take-a-hard-look-at-the-world' test.

Or maybe I'm just having a bad stats crunching day. *grumble* Ignore my moping.

See also Perpetual Postdocs and various other wonderful articles in the Chronicle of Higher Ed.
posted by redbeard at 1:06 PM on June 14, 2010 [1 favorite]

Students spend 6 or more years getting paid as little as $25,000 a year to get a PhD...

Hahahahah. As little as ~$16,000/year for a MSc and ~$18,000/year for a PhD. Even my mother feels bad for me now.
posted by hydrobatidae at 1:06 PM on June 14, 2010

I made some comment to a friend in the sciences--a Chinese national--about how the presence of so many foreign workers in the local labs must reflect the fact that American students are so disinterested in the sciences that we have a shortage of "native" scientists/technicians.

No, she corrected me, firmly. "Americans wouldn't want to do the jobs we do. Very low wages, very hard to advance."
posted by availablelight at 1:07 PM on June 14, 2010 [9 favorites]

I am sometimes not too comfortable with the choices I have made given my limited options, but I do not think I am off on my initial reaction to learning what a "post-doctorate" was. "Wait, wait, so I get a doctorate ... and then I have to go and do more stuff just so I can be one of a hundred people with this 'post-doc' competing for the same position?" Maybe I needed more faith in myself, but, damn, I do not think I could pull that off. I am not that smart, not that ambitious, and not that driven.

I started by doing research and just opening my eyes. Talked to grad students. Saw tenured professors working on things I considered Very Interesting (yeah, interesting to me, but who wants to pay for it?) driving crappy little cars that broke down a lot. Witnessed frantic scrambles for grants like a starving tribe hunting whales, including the big celebration when one was hauled in.

Universities sell paper. Very, very expensive paper. And they neglect to mention exactly what you're getting into. Decent academic positions appear to be quite hard to obtain, and merit is no guarantee. I have no idea what the solution is, but, at least in physics, somewhere around the end of your freshman year, someone ought to sit down the student in question and have a long chat about your chance of actually getting to work on cosmology, particle physics, or general relativity (what I refer to as "The Star Trek shit") are. You want a job, kid, welcome to condensed matter!
posted by adipocere at 1:12 PM on June 14, 2010 [13 favorites]

PhD in physics, now I teach high school. Lucky for me I love it. I look back at the postdoc merrygoround I nearly got on and I shiver.
posted by Shutter at 1:14 PM on June 14, 2010 [3 favorites]

It's all a cover for companies like Microsoft to hire more H1-Bs at significantly lower wages.

Can we please stop parroting this bullshit line? H1B holders don't move to some of the most expensive cities in the US because they're cheap to hire. H1B holders at companies like Microsoft or Google beat out their competition in the interview process and make the same income as their citizen or green-card holding peers. Microsoft wants (well, wanted) more H1B visas because they don't want to have to deal with limits on how and where they get their employees from. Cost has nothing to do with it.

Now, there were some abuses of H1Bs by onshore outsourcing firms, but this is hardly the majority of H1B holders and it has, for the most part, gone away now that there are established teams and facilities in places like China and India.

But this is an old canard and it's completely untrue. H1B visa holders don't make any less than their peers.

Anyway, back to the discussion at hand.
posted by GuyZero at 1:16 PM on June 14, 2010 [7 favorites]

As little as ~$16,000/year for a MSc and ~$18,000/year for a PhD.

$14,200. But that was about 8 years ago, so clearly today's kids are spoiled!
posted by madajb at 1:19 PM on June 14, 2010

But this is an old canard and it's completely untrue. H1B visa holders don't make any less than their peers.

Of course they don't. But supply and demand dictates that you can offer less money for a job when more people are applying for it. Or at the very least slow the natural inflationary increase in wages.

When Microsoft says they have a lack of people graduating with Science and Technology degrees, they mean that there are fewer people graduating in Science and Technology that are willing to settle for the amount of money they'd pay.

So, yes they don't make any less, and that's part of the problem.
posted by zabuni at 1:21 PM on June 14, 2010 [5 favorites]

where are the maverick scientists who just set up a lab in their basement and just start being scientists? like studying on their own, and making discoveries. maybe i've been reading too many sherlock holmes books lately, huh?
posted by billybobtoo at 1:22 PM on June 14, 2010

This article exactly describes life for me and my friends and co-graduates.

I'm 43; this has been going on for at least two decades. The post-doc treadmill, as we called it, was well in place by the late eighties, early nineties. I'm comparatively lucky. I managed to get into a temporary science job within a year of my doctorate and became permanent in the same job only four or five years later. My contemporaries either did the wandering post-doc / soft-money "lectureships" or left science entirely. For most of us, our late twenties and thirties were spend working on soft money in temporary terms or contracts.

Only one of my contemporaries is now a full-time tenured university prof. A few are now CxO-level people with no day-to-day research focus. Several of us are project-manager level, senior technical folks administering our own legions of young term and contract workers. Many, most?, have left science entirely.

The worrying thing is the loss of the most intellectually-important part of these people's lives. Look at the biographies of past Nobel prize winners. Most of the prize-winning work was done in their late twenties and early thirties. It's hard to be creative when your daytime job is running another set of gene plates or looking at another chromatogram for a 50-to 60-year-old research director. There's a whole generation of lost ideas there.
posted by bonehead at 1:22 PM on June 14, 2010 [13 favorites]

Universities sell paper. Very, very expensive paper.

In my experience with friends (mostly science with a few humanities thrown in), if you are paying to get your Ph.D., odds are pretty good it's not going to pay off.

If you can't find someone to pay for your science in academia, you're not going to find someone who'll pay for it in the real world.
posted by madajb at 1:23 PM on June 14, 2010 [5 favorites]

I wonder if they could start having tighter restrictions on student visas and training visas and the H1-B visa? Make it harder to obtain H1-B visas and force companies to pay H1-Bs on par with US citizens in the same position so the incentive to sponsor is only used if there truly is a shortage of American applicants? I think right now HR just has to sign something that says that they didn't receive a qualified applicant and this foreign-born person has specialized skills. But I'm not certain.
posted by anniecat at 1:24 PM on June 14, 2010

But this is an old canard and it's completely untrue. H1B visa holders don't make any less than their peers.

Some interesting links:

Government Study Find 21% Of H-1B Applications Violate Rules

The Bottom of the Pay Scale, Wages for H-1B Computer Programmers

Professor Blasts Research on H-1B Visa Workers Earning Higher Wages
posted by tommasz at 1:26 PM on June 14, 2010 [9 favorites]

Where can you still get a good job in this country? The only people I know with money to burn are egineers working for the government in some aspect, or people working with finance/accounting in some fashion.
posted by codacorolla at 1:30 PM on June 14, 2010 [2 favorites]

But this is an old canard and it's completely untrue. H1B visa holders don't make any less than their peers.

It's an old canard, but it's both true and not true. On paper, they don't make any less. But at least when I was in the software industry, the H1Bs I worked with knew damn well they couldn't push as hard for promotions, raises or transfers as their US citizen coworkers. It was always unwritten, but they knew they had to put up with just a little bit more.
posted by gurple at 1:34 PM on June 14, 2010 [3 favorites]

The academic career track for scientists has been dead for years. What's worrisome is that the private industry track is on the wane too. The biotech sector is out-sourcing more and more research to other countries, where quality is often comparable but much less expensive. The pharma pipeline has gone worryingly dry, at least in terms of clinical success rate, and patent exclusion on the blockbusters is running out. NIH is still pouring like a spigot, thanks to the stimulus, but grant competition is way up and current funding levels can't/won't last much longer. In other words, it's not a good time to be a scientist in America. The next decade's gonna be rough.

In terms of H-1B holders, there's an unspoken advantage that comes with hiring them that goes beyond pay or even skill level. They're more loyal because they have to be. If I lose my job, the worst thing that happens is that I go on unemployment. If an H-1B holder loses their job, the worst thing that happens is they get deported.
posted by dephlogisticated at 1:34 PM on June 14, 2010 [7 favorites]

where are the maverick scientists who just set up a lab in their basement and just start being scientists? like studying on their own, and making discoveries. maybe i've been reading too many sherlock holmes books lately, huh?

Well, maybe you don't have much of an idea about the sort of equipment needed for a lot of science today, or understand why so much research involves groups of people, not single maverick scientists. There's a growing DIY bio movement, but there's a lot of stuff you can't do at home, period, due to fancy expensive equipment, and a lot of techniques that involve very toxic, dangerous stuff that you can't use at home safely. I can run cheap electrophoresis gels in my basement, but can I do large-scale sequencing? I might be able to do some basic microscopy, but can I do FACS or can I do studies using P32 instead of fluorescent stuff? I can do PCR, but what about NMR, or mass spec, or ICP?

Really, once you go much beyond basic cloning and synthesis, most chem and bio projects these days involve at least a few experiments that really can't be done at home.

In my experience with friends (mostly science with a few humanities thrown in), if you are paying to get your Ph.D., odds are pretty good it's not going to pay off.

If you're paying for a science PhD, you shouldn't be getting a PhD. The system is exploitative enough when it pays grad students $25,000 stipends - and all decent programs do cover tuition and pay stipends. If you're paying, you should not be there, period.
posted by ubersturm at 1:39 PM on June 14, 2010 [6 favorites]

Where can you still get a good job in this country? The only people I know with money to burn are egineers working for the government in some aspect, or people working with finance/accounting in some fashion.

Healthcare
posted by ghharr at 1:45 PM on June 14, 2010

An intersting factor, to me at least, is the average age of university profs. I work closely with many academics on collaborative projects, probably several dozen a year. Almost all of them are in their late fifties or early sixties, ten or fifteen years older then me.

Come to think of it, I have never met a tenured academic with equivalent senority to myself---associate professor---who was my own age or younger. Some government and industry people, yes, but no academics.
posted by bonehead at 1:45 PM on June 14, 2010 [1 favorite]

I think people outside the science racket also are under the misapprehension that graduate students are "students". Maybe for the first year or so. After that, they're full-fledged scientists who just don't have a PhD yet. You know all those NOVA specials where the professor is looking into a microscope or coding on the computer or working some arcane apparatus? No. Doesn't happen (okay, it happens, but not so much). That's graduate students and postdocs. They do the vast majority of the work (professors provide advice, experience, direction, and judgement). In the US today, most scientists get paid shit until their career track permanently ends in their mid 30s to early 40s. It kind of sucks to be a low-ranking scientist in this system, but we'll all reap the "reward" for this development in the coming decades. I'm a little peeved at this myself as a postdoc, but I saw it coming and I've made sure to have marketable skills for when the hammer finally falls.
posted by Humanzee at 1:55 PM on June 14, 2010 [6 favorites]

I can do PCR, but what about NMR, or mass spec, or ICP?

We got an enormous/ancient ICP for free from the local water utility, who presumably just wanted to be rid of it. It looks like something out of Wargames and I'm pretty sure the attached computer runs MS-DOS.
posted by dephlogisticated at 1:56 PM on June 14, 2010

Aren't we talking about 2 different issues here? Microsoft has computer engineers on H1B's, not scientists. The sort of people struggling through postdocs and lack of tenured positions aren't the same people Microsoft is looking for. Is it possible that there are too many scientists but not enough engineers?

Also, regarding H1B's. When they are first hired, H1B's working for reputable companies earn salaries on par with American workers (usually better I think). However, when I had one I felt trapped because I relied on my employer for more than just a salary. Getting my greencard was a liberating experience.
posted by redyaky at 2:01 PM on June 14, 2010 [3 favorites]

I know I'm kind of idealistic about this, but isn't this part of the larger issue we have with valuing work in general, in a society where profit outweighs most other benefits?

This Curtis White essay, "The Ecology of Work" has more to say about this issue than simply why PhD holders can't make a decent living. (And yes, I'm probably a commie. )
posted by sneebler at 2:03 PM on June 14, 2010 [7 favorites]

Aren't we talking about 2 different issues here? Microsoft has computer engineers on H1B's, not scientists. The sort of people struggling through postdocs and lack of tenured positions aren't the same people Microsoft is looking for. Is it possible that there are too many scientists but not enough engineers?

Yes. And yes. If you think you want a biology PhD ask yourself really hard why you don't want a Chemical Engineering degree instead. And all the physics people I know writing software could have saved themselves a lot of trouble in interviews and had much quicker career success if they had gone into CS in the first place.

I will admit, however, to being biased in that I think anyone who isn't an engineer is a stinky poopy-head.
posted by GuyZero at 2:11 PM on June 14, 2010 [1 favorite]

Yes. And yes. If you think you want a biology PhD ask yourself really hard why you don't want a Chemical Engineering degree instead. And all the physics people I know writing software could have saved themselves a lot of trouble in interviews and had much quicker career success if they had gone into CS in the first place.

and all the CS people working in management could have saved themselves a lot of time, trouble, and debugging if they had just gone for an MBA, and why get a business degree in accounting or supply management when you can retire at 30 in finance...

until we have a society of dividend check cashers, financiers, and walmart clerks. oh wait...
posted by ennui.bz at 2:24 PM on June 14, 2010 [2 favorites]

I kinda wish that we as a society idolized scientists and the accumulation of logic and knowledge like we do sports stars, and gave them the funding that folk give to churches and televangelists.
We'd be having this discussion telepathically via neural interface with the interwebs on Jupiter, while getting our flying cars powerwashed by the local AI owned carwash.
posted by From the Fortress at 2:24 PM on June 14, 2010 [20 favorites]

To chime in with an anecdote on the other side. As a corporate employer I recently advertised for 8 months for two PhD Chemists with relevant experience. The salary was $70K/Yr in a mid-west university town that is pretty low cost to live in. For a kicker, there was a raise to $100K+ with a move to California thrown 6~12 months down the road. We utterly failed to fill the positions with post-docs. That could be for a bunch of reasons (like not wanting to move twice in a year, or not liking the mid-west etc.), but as an employer it sure didn't feel to me like there was a vast pool of qualified folks looking for work. As an aside, not a single applicate was a native born US citizen.

On a related topic an employer really has a hard time underpaying a H1B worker. You are required to pay them at least what the government defines as the prevailing wage for that position and seniority. You are also required to post their salary publicly at the job site where they work so everyone can see how much they're being paid. The only 'unfair' advantage you have is that a H1B worker can't switch jobs as easily as a permanent resident or citizen. That said they can still change jobs (via a H1B transfer to a new employer) if they really want to and they can convince the new employer that they're worth it.
posted by Long Way To Go at 2:42 PM on June 14, 2010 [1 favorite]

We'd be having this discussion telepathically via neural interface with the interwebs on Jupiter, while getting our flying cars powerwashed by the local AI owned carwash.

AI steals jobs from hard-working organics!
Support your local Human-Owned business!
posted by madajb at 2:42 PM on June 14, 2010 [4 favorites]

When Microsoft says they have a lack of people graduating with Science and Technology degrees, they mean that there are fewer people graduating in Science and Technology that are willing to settle for the amount of money they'd pay.

Just not true. I've done many interviews at both MSFT and GOOG, and we can't fill positions. There really are not enough people who are at the level we want to hire at. Other than more good candidates, the only answer would be to lower our standards. And as most programmers know, adding lower quality programmers can be counterproductive. Often you're better off just not filling the position.

Even for candidates who get past resume/phone screen, 90+% are rejected, because they do really badly in the interviews. Most interviews are laughably bad. It's really kind of depressing, and one reason I hate interviewing so much.
posted by wildcrdj at 2:44 PM on June 14, 2010 [6 favorites]

If you're paying for a science PhD, you shouldn't be getting a PhD.

Unless the offer comes with funding adequate to finish, you don't want to go there.

I think that there is perhaps a wider range in post-docs than is being mentioned here. Some are nasty, low-paid, and exploitative. Others are fine career stepping-stones, adequately paid, and totally decent places to be. Academia as a whole is now more strongly bifurcated than ever before, and that plays out at all levels, from grad student stipends (ok at well-funded schools, worse than terrible everywhere else) to post-doc pay (ditto).
posted by Forktine at 2:45 PM on June 14, 2010 [1 favorite]

$14,200. But that was about 8 years ago, so clearly today's kids are spoiled!

I made $12,000 per annum my first year out of undergraduate school (1974) with a math degree, some stat and computer programming skills working on a research project at a university. So...well, that's just sad.
posted by Mental Wimp at 3:02 PM on June 14, 2010

Not so bad
posted by Throw away your common sense and get an afro! at 3:14 PM on June 14, 2010 [1 favorite]

And another thing, when Tim Pawlenty was on the Daily Show the other night (~14mm10ss), he inferred that it was a silly waste of money for the government to support the current higher education system, because students shouldn't have to drive in from the suburbs to listen to some boring professor talk about [your favorite subject here]. Instead, there should just be an iPhone app that gives the student all the information they need.

What TPaw and his ilk forget is that academe is not just the retail store for knowledge, it's the factory. Where these pinheads imagine all that wonderful knowledge comes from I cannot imagine. Perhaps they are religious literalists and believe it springs fully formed from the minds of men, placed there by the benevolent Flying Spaghetti Monster.
posted by Mental Wimp at 3:16 PM on June 14, 2010 [2 favorites]

I'll second what Forktine said and offer another supporting anecdote/advertisement: if your dissertation has anything to do with non-equilibrium radiating gas/plasma theory or with high speed flow experiments, I know where you can work for much more than $40K. We've got a decent team put together for computational engineering, uncertainty quantification, and turbulence now, but they're still taking applications for those areas too. Oh, and if you're looking for work and you've done recent PhD work on ablation, don't be dissuaded by the fact that it doesn't correspond to an open job listing; that's not because you're not desperately wanted, it's just because nobody thinks you exist.

Still, if you were smart enough to get a PhD in any of these fields then you're hopefully wise enough to know that you could have been making more money doing something else. There are too many geeks who feel happy forgoing a bit of salary in exchange for doing more interesting work, and you don't want to have to compete in that market unless you feel the same way.
posted by roystgnr at 3:22 PM on June 14, 2010

haven't we reached a point where the degree of specialization required by the "big bucks" job openings can only be met by picking from a vastly larger global population pool...

as in: having trouble hiring from a population of of 309,000,000?

wouldn't you know it here's a guy from country x who has been doing exactly what we're looking for for 8+ years!

seems legitimate for employers to lobby for the H1B

but how to direct the 309,000,000 into impossibly narrow niches, there's the rub
posted by Hammond Rye at 3:23 PM on June 14, 2010 [1 favorite]

I want to echo Long Way To Go. Working in a government lab, we have a hard time employing foreign nationals, but it turns out that 50% or so of the post-docs that come through our building are all foreign nationals.

Also, just because you have a PhD doesn't mean you are qualified to be, or will be good at being a research scientist. There is a difference between being able to do the analysis and being the one who decides what analysis to do. The latter is a lot harder than it appears, and in my experience, most people with PhD's aren't very good at it.

Part of the problem is educating the 309 million just what opportunities do exist, before their sophmore year when they decide one hungover sunday morning to fuck it and decide to become lawyer/doctor/investment banker.
posted by spaceviking at 3:29 PM on June 14, 2010

I know where you can work for much more than $40K. We've got a decent team put together for computational engineering, uncertainty quantification, and turbulence now, but they're still taking applications for those areas too.

Experience in large-scale parallel code development, C++, and modern software engineering practices is essential. Applicants must have a Doctorate in Engineering, Science, Applied Mathematics, or a related field. This is a security sensitive position.

If you have a PhD, have extensive experience in high-performance computation and are interested in working on military research grants... we have a job for you.
posted by ennui.bz at 3:30 PM on June 14, 2010

call it "multi-disciplinary hot-spots" where commerce meets science

i was a film major man, fuck that.
posted by Hammond Rye at 3:31 PM on June 14, 2010

> Even for candidates who get past resume/phone screen, 90+% are rejected, because they do really badly in the interviews. Most interviews are laughably bad. It's really kind of depressing, and one reason I hate interviewing so much.

From what I hear from those doing interviews here this is about right. The signal to noise is just outrageous, it makes me shudder to think that some of these people might actually be able to bullshit their way into doing actual IT work.
posted by Skorgu at 3:31 PM on June 14, 2010

until we have a society of dividend check cashers, financiers, and walmart clerks. oh wait...

Well, to be fair, the original article basically posits that we have a glut of scientists. I'm all for higher education but in terms of actual job prospects there are lots of other options versus being a pure research scientist without going all the way to being a cashier.
posted by GuyZero at 3:36 PM on June 14, 2010

The signal to noise is just outrageous, it makes me shudder to think that some of these people might actually be able to bullshit their way into doing actual IT work.

There's a huge difference between "IT work" and being able to write an internet-scale machine learning system that's the backbone of a $10B annual business. 9 out of 10 candidates are probably qualified for the former. Getting 1 out of 10 for the latter is tough.
posted by GuyZero at 3:37 PM on June 14, 2010

I think the problem is that there's just not enough free-flowing capital to support the kinds of small start-ups that require good engineering/science/research. Most good science jobs are sucking off the government teet, (which I'm fine with, but it's an ever-shrinking B-cup at best these days), but forget about trying to get money from a bank. Fuckers are locked up tighter than a frog's ass.
posted by Civil_Disobedient at 3:53 PM on June 14, 2010

I know it's not what keeps everyone in the lab, but part of the deal they sell you as a grad student is that getting off the treadmill and leaving the academy makes you a second-class citizen or a failure. This is definitely true in the humanities, but I see it in my science-side friends, too. The funny thing is that leaving grad school, which was supposed to make me the bitterest person in the world, was probably one of the smartest things I ever did, and ditto for my friends who got jobs in industry on the science side.
posted by immlass at 4:05 PM on June 14, 2010 [2 favorites]

Still, if you were smart enough to get a PhD in any of these fields then you're hopefully wise enough to know that you could have been making more money doing something else. There are too many geeks who feel happy forgoing a bit of salary in exchange for doing more interesting work, and you don't want to have to compete in that market unless you feel the same way.

Science has very little ROI, and that's to be expected: what we're doing is just producing a lot of new discoveries and ideas and churning them out and maybe some of them will be really earth-shattering. But that's what we pay for: people to come up with things that might be interesting down the road, but we don't know if they're really significant until we explore them.

The problem is that this lack of interest in or understanding of ROI extends to the careers and lives of scientists: they spend weeks or months drafting grant proposals which have just a small chance of ever being funded. They engage in low-paid work over the long term, costing the American ones staggering amounts of money in opportunity costs. By contrast, the reason so many scientists are foreign born is that for them, the ROI actually pays off, as the article points out:

For scientifically trained young people from abroad, though ¡ª especially those from low-wage countries like China and India ¡ª the calculus of opportunity is different. For them, postdoc work in the U.S. is an almost unbeatable opportunity. Besides the experience and prestige of working in the world¡¯s leading scientific power, a postdoc research position is likely to pay many times more than a job at home would.

But here's the big problem that's actually a crisis: the consequence of our current structure is that most grants are becoming awarded to older-long-established Principal Investigators, following up on the same work they've been doing for decades. It sort of makes sense, because money is scarce, and grant-awarding organizations want to give their money to people who are going to be the most reliable, but it's shutting out fresh blood, few of whom are going to get a bite at the apple. We might not need more people with PhDs, but we do need more scientists getting funding for their own work, but the structure presently doesn't support that-- the money is going to a mostly similar small group of people.

I ended up striking a good balance between love of research and also enjoyment of a decent salary, and I'm happy, even though I turned down a great postdoc opportunity a couple of years ago, because the opportunity cost was just too high for me to bear. And one of the most important reasons I was able to feel good about striking that balance is because I realize that my decisions are efficient ones.
posted by deanc at 4:40 PM on June 14, 2010 [3 favorites]

The problem with science PHDs is they want to do science. Science is the single best idea that human beings ever had but for the most of scientific knowledge is non-excludable and thus exists outside the commercial market, if you can't own it it's difficult to monetize something. I am not in charge of anybody but for the average non-specialized job that MBAs are doing (most stuff that isn't finance) I would rather hire a scientist that could write clearly. I don't really care if the job has nothing at all to do with their specialization (though I imagine the prospective scientist would care). I think the amount of cognitive horsepower involved in the science and mathematics is far higher than in a typical MBA program. I would be able to get top shelf intellects at rail prices.

Maybe there is something I'm missing but the good MBAs I've met were just good because they were smart people. They never seemed to display any specialized knowledge that they picked up in school. They never seemed to be particularly good at things that require an MBA education (whatever that is).
posted by I Foody at 4:41 PM on June 14, 2010 [2 favorites]

they mean that there are fewer people graduating in Science and Technology that are willing to settle for the amount of money they'd pay.

If those lousy idiot Americans aren't willing to accept close to USD$100k per year for a job straight out of their undergrad degree then fuck them. I mean, what the hell do the precious little snowflakes think they're worth?

Yes, I am a H1B worker at Microsoft, and yes if you happen to be the next one to say 'couldn't they find any Americans to do it?' I'll spit at you. Because actually no, I'm assuming they couldn't. If they could, they probably would have been happy to increase that salary by several grand a year in lieu of spending another $20,000 to interview me and relocate me from Australia when I'm a completely inexperienced college kid.
posted by jacalata at 5:17 PM on June 14, 2010 [4 favorites]

There's a huge difference between "IT work" and being able to write an internet-scale machine learning system that's the backbone of a $10B annual business. 9 out of 10 candidates are probably qualified for the former. Getting 1 out of 10 for the latter is tough.

And yet I interview candidates for a somewhat to significantly less particular company, and I would say my ratio of interviews-to-accepts is also around 90% (maybe a little better for new college grads and contract employees). It is shocking how many people out there working in "IT" cannot. write. code. At. All. When you find yourself asking people claiming 10 years of development experience to write what amounts to a simple for loop, and they are unable to do it, you know things are bad.
posted by ch1x0r at 5:18 PM on June 14, 2010 [4 favorites]

When you find yourself asking people claiming 10 years of development experience to write what amounts to a simple for loop, and they are unable to do it

Heh, one of my interviews at MSFT I asked a guy a simple question (something like removing an element from a linked list). He struggled right off the bat, etc. Eventually he said "oh shit, I have to use a loop". Which of course he couldn't do...
posted by wildcrdj at 5:26 PM on June 14, 2010

Is it possible that there are too many scientists but not enough engineers?

The problem is that no one wants to hire scientists to be engineers, even though scientists generally have more than enough intellect and creativity to be successful in engineering, and often even better writing and speaking skills than their engineering counterparts. But it may take them six months, even a year or so to ramp up. They're not fresh college grads to begin with, and to make training them worthwhile, you've got to keep them around until they're at least 30 or 35 -- maybe 40! -- and getting saddled with all these inconvenient family obligations and possibly medical problems.

Or maybe the real problem is that some of them get flustered when you ask them to do some pointer arithmetic in an interview.
posted by transona5 at 6:01 PM on June 14, 2010

I made $12,000 per annum my first year out of undergraduate school (1974) with a math degree, some stat and computer programming skills working on a research project at a university. So...well, that's just sad.

Well, it wasn't so bad. Cheap mid-western town, so expenses were low.
Definitely just a touch over minimum wage though.
posted by madajb at 6:05 PM on June 14, 2010

The problem is that no one wants to hire scientists to be engineers

There's some truth to this, but it is tough to tell who will make a good engineer. Let's talk about programmers, since that's what I know about. In school (Caltech, so everyone was a scientist or an engineer, thats all we had) there were plenty of scientists who struggled with CS1. And of course many who did well. But just like with science, there's a combination of learning and innate ability that can be hard to judge without experience. I can study physics all I want, but I don't grok it enough to be good at it the way many of my classmates were.

That being said, odds are certainly much higher for a good scientist than a person at random. Math & logic are prerequisites that most scientists should be pretty good at.

But given the issues with hiring, the large companies absolutely should be trying to do more training of promising non-engineers. Details of how long you would try before giving up are tough to decide on, but it's something that could be experimented with.
posted by wildcrdj at 6:14 PM on June 14, 2010

This is just like the market for graduating law students. Why is anyone surprised?

What Admiral Haddock Forktine said. The expected economic payoff from any professional degree (JD, MBA, PhD) is negative, except possibly for the very top-tier schools. Going into debt in pursuit of a negative expected payoff is stupid.

Disclosure: I'm a tenured CS professor.
posted by erniepan at 6:16 PM on June 14, 2010 [1 favorite]

This is just like the market for graduating law students. Why is anyone surprised?

I'm pretty sure we already had this exact thread before except with English PhDs last time.
posted by GuyZero at 6:19 PM on June 14, 2010

Back in the 80's we undergrads speculated that positions would be opening up once the then current crop of professors retired. Ergo, it was a good idea to attend grad school and get a PhD. Which I started to do, for one quarter and then flamed out. Recently I spoke to a friend who had returned to my alma mater in his 50's as a postdoc. Many of the same tenured professors from 25 years ago are still there and show no signs of stepping down anytime soon.
posted by telstar at 6:31 PM on June 14, 2010

I don't agree with the idea that we have too many scientists. Sure, we have too many to all become principle investigators. You can't just slim down the current system by firing all the postdocs and having fewer grad students though ---no work would get done. The problem is that the current employment profile has been coupled with a system designed to pay non-faculty scientists just enough to survive on, with the opposite of job security (guaranteed that postdoc ends, you can't get promoted from it, you MUST leave). This state gets dragged out over a long period of time, so you have scientists working over a sizable fraction of their lives, making crap wages and having little prospect for for improvement.

The thing is, I like my job, as do most other postdocs I know. It's the career situation that sucks. So if you can improve the living situation of graduate students and postdocs to be commensurate with the importance of their role in the scientific process (or at least make it a little less shit) then problem solved. The real issue is that like with every publicly supported enterprise, everyone wants to do things cheaply. So the framing "too many scientists" is merely presuming the ultimate solution: fewer scientists in the US. The system won't change much, we'll just have fewer labs, and maybe pay will bump up a little. All those foreign scientists who come here to train are going to go back home and take their expertise with them. In a generation, the scientific preeminence of the US will retreat to background. Not the end of the world, although in my opinion it's a mistake.

On preview: in science, grad students and postdocs generally don't go into debt. Also, most eventually leave academia and get decent jobs. Depending on the field, their employment prospects can be very good. They just get a lengthy self-imposed delay on their earning potential. I've never met anyone who went into science because they thought it would make them big money though. Pretty much everyone is in it because they think it's interesting, worthwhile, and important. So we're not talking about a failure to correctly calculate ROI. With a PhD in science, the school doesn't matter, only who you work with.
posted by Humanzee at 6:32 PM on June 14, 2010 [2 favorites]

I'm pretty sure we already had this exact thread before except with English PhDs last time.

Yes, and both of them have made me feel better about taking my BA and running. I've often wondered if I should have gone into the sciences instead; guess not.

With this huge apparent glut of educated people, I really wish we'd take a WPA approach, use govt money to hire them to work on Big Problems (like alternate energy, agriculture, pollution, education, etc) and get some badly-needed shit done. It's more than stupid to have all these capable knowledge-filled people pushing around insurance spreadsheets or what have you because that's the only work they can get.

But then we might have to divert our defense spending or have fewer wars or some commie shit, so fuck that.
posted by emjaybee at 6:48 PM on June 14, 2010 [4 favorites]

The proliferation of post-doc requirements is striking. I've been really surprised given the overall supply how bad some candidates who make it to giving a job talk are. Particularly in lab sciences, hiring committees want real proof that someone can think on their own and wasn't just a vehicle for their PI. The issue of oversupply of grad students with no hope has been discussed for a long time. The plot of age at first R01 in that last one is demoralizing.
posted by a robot made out of meat at 6:57 PM on June 14, 2010 [2 favorites]

The expected economic payoff from any professional degree (JD, MBA, PhD) is negative, except possibly for the very top-tier schools.

If you specialize in Medieval Russian snail painting, then, yeah, you get what you paid for.
posted by madajb at 8:11 PM on June 14, 2010

If you have a PhD, have extensive experience in high-performance computation and are interested in working on military research grants... we have a job for you.

A reasonable conclusion, but an incorrect one. The physical application is atmospheric entry, but for crew capsule geometries, materials, and trajectories. The trouble is that a lot of NASA stuff (and NASA-quality stuff) is export-controlled, because of the dual-application fear with regards to warhead reentry.

There is sensitive research out there that isn't military research. Civilian nuclear power, to pick another example hiring computational mathematicians right now. And even military research isn't necessarily weapons development research. I hear that ARPANet thing turned out to have some minor civilian applications. To pick an example I ran into more recently: they still make non-US citizens jump through lots of hoops to work with the Army Corps of Engineers, even for applications like the science of flood prediction which are about as non-dual-application as you can imagine.
posted by roystgnr at 8:43 PM on June 14, 2010

*cracks whips at scientists*

Muuahaahahaaa! Where are your jokes about arts degrees and toilet paper now, underlings?
posted by obiwanwasabi at 1:30 AM on June 15, 2010 [2 favorites]

But then we might have to divert our defense spending or have fewer wars or some commie shit, so fuck that.

I wonder how many scientists and engineers are paid through defense spending.
posted by anniecat at 6:42 AM on June 15, 2010

They like to suck in as many potential candidates as they can. Then they skim what they themselves call "the cream of the crop" and throw the rest out to sink or swim.

In the mid-60s the government declared a need for 20,000 new scientists within 10 years. 6 years later, the Apollo program ended, and 45,000 scientists and engineers with valuable high-end experience were thrown into the marketplace.

Employers still got what they needed. Guess what the graduates got?

Students: NEVER EVER pay any attention to what they say. Study what you really REALLY WANT to study. NOBODY knows the future.
posted by Twang at 7:48 AM on June 15, 2010 [8 favorites]

This kind of crazy job prospects in the US and Australia was one of the reasons it was so easy to move to Mexico. Just finished a phd, and a postdoc later I have the equivalent of a tenured position, still doing the kind of science I always did. Granted as a behavioural ecologist, working with spiders, its a lot easier since most of my equipment is really basic and my animals are easily found, and I'm practically living in an area surrounded by my field site :)
posted by dhruva at 11:19 AM on June 15, 2010

I wonder how many scientists and engineers are paid through defense spending.

The DoD spent $80.7 billion in FY 2009 on R&D alone. Lest you imagine that goes solely towards the creation of shiny new death machines, I can personally attest that our research on traumatic brain injury and PTSD are both bankrolled entirely by the DoD. Indeed, without those grants our nonprofit lab might have folded during the recession.

Although, from a political standpoint, I'd much rather have NIH ($29.5 billion FY 2009) or NSF ($6.85 billion FY 2009) dispensing those funds, I'm damn glad that the money is there and going towards some good causes.
posted by dephlogisticated at 11:26 AM on June 15, 2010

Even for candidates who get past resume/phone screen, 90+% are rejected, because they do really badly in the interviews. Most interviews are laughably bad. It's really kind of depressing, and one reason I hate interviewing so much.

Presume you were not expecting a slick-ass MBA sales-job interview from the nerds?
"My only faults is that I work too hard!" -Trainspotting.
posted by ovvl at 8:48 PM on June 15, 2010

Anyone capable of operating a time-domain reflectometer is capable of learning how to behave in an interview. People have poor interview skills because they're never really taught.
posted by GuyZero at 9:30 PM on June 15, 2010

And in my case of having tons of horrible interviews, it's the technical skills that are lacking, not the interview skills.
posted by flaterik at 9:55 PM on June 15, 2010

I got a PhD in microbiology in the US and moved to Australia for my post-doc. Overall, the level of funding for research science in Australia is much, much lower than it is in the US. However, they actually pay research staff decent wages- my salary as a first year post-doc is double what I would make in the US. Here, we reuse weigh boats and rack our own tips to save money, whereas in the US labs will easily drop 500K on a piece of equipment while paying a highly trained PhD peanuts to run it. I don't know if there's a way to make reasonable salaries more of a priority in american academic science, but I wish that there were.
posted by emd3737 at 11:21 PM on June 15, 2010 [1 favorite]

I'd agree that emjaybee proposal is basically the only realistic option, people need to be put to work doing useful stuff, which means spending more money on science and technology. For mathematics, I'd suggest largely replacing the NSF grants system by explicitly postdoctoral grants with several properties :

(1) Grants are awarded based solely upon perceived intellectual ability and research domain, last 10 years, and are completely transferable between institutions.

(2) Grant recipients working inside academia receive $60k per year with raises and minimal travel/conference budget. Grant recipients working outside academia receive merely a $5k per year salary bonus, but the remainder could be converted into start up funds for "academically interesting" companies.

(3) Academia hosting institutions receive $3k, $6k, $12k per year depending upon whether the grant recipient is non-tenure-track, tenure-trank, or tenured, respectively. If teaching replaces any portion of an applicants salary, that portion immediately becomes available for travel, conferences, and equipment, although now the institution may claim overhead when the money is spent.

(4) All grants awards are associated with a scientific committee of three tenured faculty at three different academic institutions, one must be U.S. based. Any particularly dramatic alterations like changing institutions must be approved by the committee. Any committee changes must be approved by the NSF.

In short, you fix the problem by tying all the money to the postdocs that actually do the work, which incidentally gives their scientific committees the freedom to actually do science, instead of writing grants all day.
posted by jeffburdges at 8:01 AM on June 16, 2010

The level of funding for research science in Australia is much, much lower than it is in the US. However, they actually pay research staff decent wages...

The same is true in Canada. I was shocked to learn what passes for post-doc pay in the US. Even at places like NIH and NIST, rates are about 1/2 to 2/3 of what I pay. My particular pay range is required by a public program, but that's true for most post-docs in Canada.

To be clear, it's not that being a post-doc is so bad. These jobs are often quite reasonable pay for first jobs and can offer a great deal of job satisfaction. The problem is that there's no place to go after the post-doc, and the system is geared to flip these jobs every year or two.

An intermediate level between post-doc and PI would solve the "journeywoman/man" problem that the 20 and 30 year-olds have. Granting bodies have no place for such though; you're either a post-doc or a PI. Research assistant positions are scarce as hen's teeth.The only place they really exist is in Federal labs, in both Canada and the US.
posted by bonehead at 10:56 AM on June 16, 2010