About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« One and Done | Main | Gitcher SF5 Groups Right Here »

April 17, 2014

Changing A Broken Science System

Email This Entry

Posted by Derek

Here's a suggestion for a total reform of the graduate student/postdoc system of scientific labor and training. It's from a distinguished list of authors, and appears in a high-profile journal, and it says without any equivocation that the system we have is in major trouble:

In the context of such progress, it is remarkable that even the most successful scientists and most promising trainees are increasingly pessimistic about the future of their chosen career. Based on extensive observations and discussions, we believe that these concerns are justified and that the biomedical research enterprise in the United States is on an unsustainable path. . .We believe that the root cause of the widespread malaise is a longstanding assumption that the biomedical research system in the United States will expand indefinitely at a substantial rate. We are now faced with the stark realization that this is not the case. Over the last decade, the expansion has stalled and even reversed.

They trace the problem back to the post-World War II funding boom (Vannevar Bush's "Endless Frontier"). I have to say, the paper gives the impression (no doubt for lack of space) that the progress of funding in the biomedical sciences was smoothly upwards up until about 1990 or so, but as I understand it, the real kick was the post-Sputnik expansion. The 1960s were the real golden years for federal science and education spending, I think, as witness the profusion of buildings from that era to be found at many public universities. You can spot them from a hundred yards away, and boy, are there are lot of them. The authors lump that era in with the 1970s, but that latter decade, at least post-1973 or so, was hardly a period of a "vibrant US economy", as stated.

The doubling of the NIH's budget is also dealt with like a matador deals with a bull - a flick of the cape. But there's no doubt that the situation now isn't good:

However, eventually, beginning around 1990 and worsening after 2003, when a rapid doubling of the NIH budget ended, the demands for research dollars grew much faster than the supply. The demands were fueled in large part by incentives for institutional expansion, by the rapid growth of the scientific workforce, and by rising costs of research. Further slowdowns in federal funding, caused by the Great Recession of 2008 and by the budget sequestration that followed in 2013, have significantly exacerbated the problem. (Today, the resources available to the NIH are estimated to be at least 25% less in constant dollars than they were in 2003.)

The problem has been the same one faced by highway engineers: double the lanes on the highway, and new traffic fills up it again. Extra NIH money has been soaked up, and more, by an expansion in the customers for it. Even if their history is a bit off, the authors' analysis of the current situation seems to me to be right on target. :

The mismatch between supply and demand can be partly laid at the feet of the discipline’s Malthusian traditions. The great majority of biomedical research is conducted by aspiring trainees: by graduate students and postdoctoral fellows. As a result, most successful biomedical scientists train far more scientists than are needed to replace him- or herself; in the aggregate, the training pipeline produces more scientists than relevant positions in academia, government, and the private sector are capable of absorbing.

The result, they say, has also been Malthusian: an increasingly nasty competition for resources, which is taking up more and more of everyone's time. It's creating selection pressure favoring the most ruthless elbow-throwers and body-slammers in the bunch, and at the same time making them scientifically timid, because the chances of getting something unusual funded are too low. (Paula Stephan's thoughts on all this are referenced, as well they should be). You may now see the birth of the "translational research" bandwagon:

One manifestation of this shift to short-term thinking is the inflated value that is now accorded to studies that claim a close link to medical practice. Human biology has always been a central part of the US biomedical effort. However, only recently has the term “translational research” been widely, if un- officially, used as a criterion for evaluation. Overvaluing translational research is detracting from an equivalent appreciation of fundamental research of broad applicability, without obvious connections to medicine.

I'm not quite so sure about the evocations of the golden age, when great scientists were happy to serve on grant review committees and there was plenty of time for scientific reflection and long-term thinking. I would place those further back in history than the authors seem to, if they existed at all. But there's no need to compare things today to some sort of ideal past - they're crappy on the absolute scale, prima facie.

From the early 1990s, every labor economist who has studied the pipeline for the biomedical workforce has proclaimed it to be broken. However, little has been done to reform the system, primarily because it continues to benefit more established and hence more influential scientists and because it has undoubtedly produced great science. Economists point out that many labor markets experience expansions and contractions, but biomedical science does not respond to classic market forces. As the demographer Michael Teitelbaum has observed, lower employment prospects for future scientists would normally be expected to lead to a de- cline in graduate school applicants, as well as to a contraction in the system.
In biomedical research, this does not happen, in part because of a large influx of foreign applicants for whom the prospects in the United States are more attractive than what they face in their own countries, but also because the opportunities for discovering new knowledge and improving human health are inherently so appealing.

Too many players have an incentive to act as if things are supposed to go on the way that they have - universities get overhead out of grant money, so why not hire as many grant-bringers as possible? And pay salaries, as much as possible, out of those grants instead of from university funds? Why not take in as many graduate students as the labs can hold? The Devil is (as usual) on hand to take the hindmost.

The rest of the paper is an outline of what might be done about all this. The authors propose that these steps be phased in over a multiyear period, with a goal of making funding more sensible (and predictable), and altering the way that the academic research workforce is recruited and handled. Here are the steps, in order:

1. Require longer-term budgeting for federal research funding.

2. Gradually reduce the number of PhD students in the biomedical sciences. Support them on training grants and fellowships rather than out of research grants. The rules barring the funding of non-US citizens through these routes need to be changed, because these should become the only routes.

3. Make more funding opportunities available between science career paths and allied fields, so that there are more possible off-ramps for people with science training.

4. Gradually increase the salaries offered federally-funded post-docs, so the system doesn't overload with cheap labor. Limit the number of years that any postdoctoral fellow can be supported by federal research grants, and require salaries to be at staff scientist level if the person continues after this point.

5. Increase the proportion of staff scientists. Universities and granting institutions need to be given incentives to value these positions more.

6. Change at least some of the NIH granting mechanism to a system more like the Howard Hughes fellowships - that is, award longer-term money to outstanding people and labs, rather than to individual proposals. There should be several separate programs like this for different career stages.

7. Set aside a higher proportion of grants for "high-risk, high-reward" ideas.

8. At the same time, consider capping the total amount of money going to any one group, because of the diminishing-returns problem that seems to set in past a certain level.

9. Make grant evaluations less quantitative (number of publication, impact factors) and more qualitative. Novelty and long-term objectives should count more than technical details.

10. Broaden the reviewing groups (in age, geographical representation, and fields of expertise) to keep things from getting too inbred.

11. Start revising the whole "indirect cost recovery" system for grants, which has provided perverse incentives for institutions, with special attention to paying faculty salaries out of grant money.

The authors note that all these changes will tend to increase the unit cost of academic research and shrink research group sizes, but they regard these costs as worthwhile, because (1) the current system is artificially propped up in both regards, and (2) the changes should lead to higher-quality research overall. A lot of these idea seem sound to me, but then, I've never had to deal with the academic research environment. There will, I'm sure, be many people who look on one or more of these proposals with dismay, for various reasons. It will be quite interesting to see if this gets any traction. . .

Comments (58) + TrackBacks (0) | Category: Academia (vs. Industry) | Graduate School


1. neurologist649 on April 17, 2014 8:36 AM writes...

I'm reminded of the short story written in 1961 by Leo Szilard called "The Mark Gable Foundation", wherein many aspects of this mess were presaged...

Permalink to Comment

2. on April 17, 2014 8:47 AM writes...

Here is the problem. They may be very distinguished scientists, but they are lousy central planners with proven track record of being lousy. Instead of glorifying them, we should let everyone know that the current problems are due to excessive central planning in science, and cure cannot be the same. My suggested cure is discussed here.

I covered this topic many many times in my blog. In its heyday, US and British academia used to be completely decentralized, where wealthy donors paid to the universities and then university heads spent the money for internal developments. To make sure donors did not have any say over how the money is spent, tenure system was installed. It was not to give permanent jobs to the academics, but to make sure they could freely criticize the donors without worrying about funding getting cut.

Today, the government has established monopoly over the funding and even worse, every academic I know (tenured or untenured) is afraid of criticizing their source of money fearing retribution. So, you can see very little talk about endless wars or NSA or anything else (ENCODE) among supposed 'intellectuals'.

Permalink to Comment

3. Anonymous on April 17, 2014 8:52 AM writes...

See other writing on the topic:
Paula Stephen's book on funding science

Micheal Teitlebaum's "Boom Bust" book on scientific workforce

Even The Emperor of all maladies covers the "War on Cancer" expansion

Legislation/organizations to fix the issue (in parts):
H.R. 4384 America HEALs Act

NPA (National Postdoctoral Assn)

Unionization of graduate students if the collegiate football players unionization stands

Permalink to Comment

4. jred on April 17, 2014 8:54 AM writes...

As a (very) recent Ph.D. grad who thankfully landed in industry, it's nice to see people trying to look at the problems we have. Numbers 4, 5 and 7 on that list are badly needed.

Another thing we need to start doing is failing people out of graduate school. The problem is much worse now than even 5 years ago (when I started), I can only imagine how things have degraded from a decade or so ago (a fact that our adviser would beat us over the head about once a year "you guys have it so much easier than I did").

A fair percentage of domestic students are starting to use graduate school (and the available cheap student loans) to extend their fun from undergrad. Meanwhile the international students come in and show them up. There's no passion from the domestics, and no ramifications for their laziness (at least at my Big 10 school).

Permalink to Comment

5. Anonymous on April 17, 2014 9:13 AM writes...

I would agree with the idea of reducing # of PhDs awarded. If that means failing out more or just not accepting as many, so be it. As someone mentioned, grad school is often viewed as an easy way to keep loans in deferment while getting paid a stipend, which isn't a ton of money, but you can eat and buy beer. I did grad work at an SEC school and this was quite common. International students in general far out performed their domestic counter parts. In my program we currently have 4 students who are 10+ years into program, how does that happen??!!

Permalink to Comment

6. Anonymous on April 17, 2014 9:13 AM writes...

I would agree with the idea of reducing # of PhDs awarded. If that means failing out more or just not accepting as many, so be it. As someone mentioned, grad school is often viewed as an easy way to keep loans in deferment while getting paid a stipend, which isn't a ton of money, but you can eat and buy beer. I did grad work at an SEC school and this was quite common. International students in general far out performed their domestic counter parts. In my program we currently have 4 students who are 10+ years into program, how does that happen??!!

Permalink to Comment

7. Anonymous on April 17, 2014 9:14 AM writes...

I would agree with the idea of reducing # of PhDs awarded. If that means failing out more or just not accepting as many, so be it. As someone mentioned, grad school is often viewed as an easy way to keep loans in deferment while getting paid a stipend, which isn't a ton of money, but you can eat and buy beer. I did grad work at an SEC school and this was quite common. International students in general far out performed their domestic counter parts. In my program we currently have 4 students who are 10+ years into program, how does that happen??!!

Permalink to Comment

8. Wavefunction on April 17, 2014 9:17 AM writes...

I think the point about supporting students on training instead of research grants is the single most important point in that piece. Otherwise there's no way the system will stop using grad students and postdocs as endless, cheap fodder for publications and funding.

Permalink to Comment

9. Anonymous on April 17, 2014 9:21 AM writes...

If there are too many science PhD's, then what is the point of graduate fellowships, especially ones funded by the federal government? NSF graduate research fellowship program especially- get rid of it and use the money for something else.

Permalink to Comment

10. Anonymous on April 17, 2014 9:27 AM writes...

This all sounds like a severe STEM shortage to me.

Permalink to Comment

11. Anon on April 17, 2014 9:28 AM writes...

I would love to see this get traction, but it likely won't. Francis Collins blew off the Biomedical Workforce Task Force every time they've raised an issue. He absolutely refuses to acknowledge their findings.
The most he has ever done is request a follow-up investigation which delays everything another year and makes a public comment about the issue being important but not substantial.

Permalink to Comment

12. The Iron Chemist on April 17, 2014 10:00 AM writes...

With respect to number 7, they already tried that with the NIH R21 proposals. The problem was that too many of the referees simply didn't play along and reviewed all of the R21s as if they were R01s. With most grants, you worry about shifting goal posts from one revision to another. With the R21s- to extend the analogy- you worry about whether you're even in the right stadium.

The staff scientist suggestion is a fantastic one, but institutions will not go for it. Not when they're quickly replacing their faculty with sundry cheaper replacements (adjuncts, lecturers/instructors, etc.). They're paying substandard wages to overqualified folks in return for three sections of general chemistry per semester. They won't pay someone to just do research unless a bigwig applies a lot of leverage.

The before-mentioned points about the selectivity of the PhD programs are spot-on. Even now, there are too many students who view graduate school as Undergraduate v. 2.0 and exit the programs with BS-level skills (or worse).

Permalink to Comment

13. Skeptic on April 17, 2014 10:08 AM writes...

I was in grad school when the US economy bottomed out. We had some tough recruiting years after that. I remember my PI coming into the office one day and yelling at all of us that the incoming class was smaller than they'd had in a long time and it was because we weren't enthusiastic enough during recruitment weekend. At the time the school had just lost a significant amount of state funding and was having to trim the fat off the budget already. Additionally, his own grant situation was dire and to my knowledge he has not secured any new grants since I left.

None of the current financial realities have penetrated his consciousness. The school should still be recruiting at a relentless pace, no matter how much funding gets cut, no matter how few grants there are, no matter how horrid the job market is afterwards. The scary thing is that he's not alone in this thinking. Not by a longshot.

Permalink to Comment

14. anonymouscb on April 17, 2014 10:15 AM writes...

While I agree with many of the authors' solutions to the current crisis in grant funding, several of the ‘solutions’ are either problematic or unfair including the suggestion that graduate students should be supported by training grants and fellowships rather than research grants. This ‘solution’ will undoubtedly hurt those beginning their careers because one of the criteria for obtaining an NIH fellowship is the qualifications of the mentor, which includes how many people the PI has placed in academia (industry does not appear to count).
Also the suggestion that we focus on hiring more staff scientists is a non-starter at least for me because I like to have turnover in my group so that we can bring in fresh ideas and techniques into my lab. This solution will also have the perverse effect of leading to unemployment for the senior scientists when lab funding runs out or the PI retires (ok dies is more likely).
Having served on numerous study sections and having submitted both successful and unsuccessful grants, I am confident that all proposals are reviewed qualitatively because if they don’t ‘like’ your grant (for whatever reason technical or otherwise) the reviewers’ will come up with a reason/excuse not to fund the grant. So, I am not sure how making the reviews more qualitative would improve the situation. The real problem here is that funding is so tight right now that even the ‘big shots’ in science are starting to feel the pinch.
I would note, however, that I agree with the authors’ belief that more qualified reviewers are needed for the funding panels and that we should limit the need for widespread geographical representation (Derek, I don’t believe they advocated this). We just need the best people. I also agree that the panels are often too inbred, but panels that have too broad a scope also tend to become parochial and the different subgroups on the panel still have a tendency to pick winners and losers within their own field.
All this being said, I do agree that the NIH should change its granting mechanisms and award more funds along the lines of Howard Hughes, where you invest in individuals rather than specific projects. However, the current numbers of people funded by these mechanisms is too small and should be expanded dramatically. Also, it makes no sense to review hundreds of submissions if you are only going to fund I also agree that NIH should revisit the "indirect cost recovery" system for grants to abolish the perverse incentive for institutes to hire individuals but have no ‘skin in the game’ when it comes to paying their salaries (except in providing bridge funding between grants). This change alone will rapidly shrink the scientific workforce.
One other solution that I agree with and that Derek does not mention, in his otherwise wonderful summary of the editorial, is the need to sunset the big science projects that the NIH currently funds (e.g, GWAS studies, the structural genomics initiative, etc). These big projects are diverting funds from individual labs and if they were sunset would free up monies to support these labs, assuming that is a priority of the NIH.
Finally, NIGMS recently noted that investigator initiated grants funded by that agency have fallen from 99% to 80% and that this decrease correlates with an increase in the number of targeted applications (i.e. requests for proposals). Although sometimes it is necessary to focus on a particular area of science and devote resources to it, too often it seems that these RFPs are being pursued because someone with connections can convince a program officer to write an RFP or, even worse, a program officer believes that they know what the next big thing in science is.

Permalink to Comment

15. Anonymous on April 17, 2014 10:24 AM writes...

I was laid off in 2009 after the 2008 downturn from pharma.

Went to grad school and am now in a lab getting rocked by funding cuts and we still haven't been able to secure funding after almost a dozen grants.

The problem is that biomedical science is inherently unstable. STEM =/= biological/medical/chemical sciences. Anything biomedically related is always going to have terrible job prospects and job stability.

I can't tell you how absolutely infuriating it is to propose a R21 for EXPLORATORY WORK on a very novel idea, only to have it immediately shot down by the NIH because "there isn't enough data" or "it is unclear that this will be disease modifying". Well isn't that the point of the stupid grant, to get the money that will allow us to do that kind of work?

Permalink to Comment

16. Anonymous on April 17, 2014 10:26 AM writes...

Off Topic:

Is anyone else experiencing the front page not loading correctly? It cuts off in the middle of a sentence and the right column doesn't load at all. From the source, it's not a display issue but the page simply didn't download completely.

Permalink to Comment

17. chemistry prof on April 17, 2014 10:42 AM writes...

In reducing the number of PhDs, entire graduate programs should disappear.

There are far too many mediocre places (and mediocre PIs) that issue PhDs, and there is enough sentiment in the grant reviewing process that money should be spread around to an extent. For example, it is much harder to get a 2nd NIH or NSF grant than a 1st, even if the grants score equally well. I have served on several grant review panels with this bias.

This sentiment is less damaging when funding rates are high. It is going to be a bigger and bigger problem in an era of reduced resources, especially when people forget that federal funding represents a taxpayer investment, not a welfare program. Entire low-end doctoral programs should disappear (or perhaps become masters programs that feed into highly competitive PhD programs).

I also agree with the sentiment that standards for remaining in a PhD program should be higher. I have seen many students who should have left in year 2, often because research does not remain their main interest, but stick it out because of the stigma of quitting. Once these students progress several years into the program, PIs view these people largely as sunk costs that need to graduate to clear out of the system. This practice doesn't benefit anyone and becomes even less OK in an era of belt tightening. A better practice would be to have a hard look at progress and interest in year 2 before a student becomes a true PhD candidate.

Permalink to Comment

18. Justin on April 17, 2014 10:50 AM writes...

Grad students and post-docs are fodder, with for the most part no real training or teaching for a "real job." There are absolutely too many students - students that have no business being given (not earning) a PhD - and too many students that can reasonably be taught and trained by the PI. Reduce lab sizes and there is more grant money to go around, with good people doing good work. The current "system" is not sustainable.

I may come across as bitter, but I'm really not. Looking back (~10 years) I had an overall good experience - but it could have been better.

Permalink to Comment

19. MarkP on April 17, 2014 10:50 AM writes...

Re: anonymouscb

"Also the suggestion that we focus on hiring more staff scientists is a non-starter at least for me because I like to have turnover in my group so that we can bring in fresh ideas and techniques into my lab."

Sure. It's the fresh ideas, not the extremely low salaries. And what, after a few years in your lab their ideas and techniques become stale? Kudos though for at least being honest. Most PIs claim that they are training their grad students/postdocs, making them much more valuable.

Permalink to Comment

20. watcher on April 17, 2014 11:17 AM writes...

Yes, there are a lot of problems with the current system.

But, I can't see there will be big changes in the system, as there are too many influential people who'd have to give up too much throughout the entire academic and funding organizations.

Permalink to Comment

21. rico on April 17, 2014 11:18 AM writes...

11. Start revising the whole "indirect cost recovery" system for grants, which has provided perverse incentives for institutions, with special attention to paying faculty salaries out of grant money.

this should be #12, #13, and so on….

Fiscal 2011, the federal government (read: Taxpayers) handed out $40 billion for R+D to universities. 896 Universities total received money AND just 10 of those universities received 20% of the total. University funding mechanisms need complete structural reform (including tuition, tax status, endowments, etc). How likely is this to happen in our life times? They have entire infrastructures, departments, administrations just to maintain it…Universities are thinking "we've got to feed the beast."….they are not thinking "what's the most sustainable system to train students and teach new knowledge to students".

Top Universities in 2011:
10. Duke. $585 million in federal dollars (taxpayers). 57% of R+D comes from federal. endowment $5.56 billion.
9. Wisconsin - Madison. $594 million in federal dollars. 53% of total from federal. endowment $1.8 billion
8. UCSD. $637 million federal. 63% total from federal. endowment $371 million.
7. Columbia U. $645 million federal. 73% total from federal. endowment $7.6 billion.
6. Stanford. $656 million federal. 72% total from federal. endowment $17 billion.
5. UPitt. $662 million federal. 74% total from federal. endowment $2.62 billion.
4. UPenn. $707 million federal. 80% total from federal. endowment $6.75 billion.
3. U Michigan. $820 million federal. 64% total from federal. endowment $7.69 billion.
2. UW-Seattle. $949 million federal. 83% total from federal. endowment $2.1 billion.
1. Johns Hopkins. $1.88 billion federal. 88% total from federal. endowment $2.6 billion.

A majority of JHU money comes from Defense and NASA, not HHS - this is an outlier.

Total: $8.135 billion from federal gov.
across these schools, on average, 70% of EACH YEAR's R+D budget comes from federal dollars.
These schools have collectively: $54 billion in endowment (that they continue to invest, tax free or generally v low taxes).

Of course this is unsustainable. Using an indirect cost average of 70%, this is about $3.35 billion JUST FOR INDIRECT COSTS such as heat, lighting, building maintenance, administrative salaries, etc.

What rational organization builds a system that REQUIRES that 70% of the yearly operating budget comes from a fickle source like the government? This is subject to market forces, politics, elections cycles - just like everything else.

Permalink to Comment

22. p on April 17, 2014 11:21 AM writes...

One thing overlooked in the discussion is that grad students aren't just fodder for PIs. Colleges need TAs. Some need a lot. The demand for undergrad labs is very high. Cut the number of grad students and put them all on training grants and someone will have to teach those labs.

Now, there are people who could do it. But it would cost more than it costs now and collges are trimming like crazy.

Permalink to Comment

23. Anonymous on April 17, 2014 11:33 AM writes...

A great discussion of an important article. I have 5 thoughts to add on the matter:

1) The short-term funding contraction may be part of the immediate problem, but it must be stated that the greater problem cannot be solved by further expenditure. NIH funding (inferred from those 5 institutes which have existed continuously) has increased 33x in inflation-adjusted dollars since 1955. More funding now will provide a short-term solution but a long-term headache.

2) The article is correct to point out that the principle issue is that the labour market is saturated with PhD holders. There are about 10-20 times as many basic scientists working today as there were in the 1950s. To reduce the number is doubtless desirable, but almost impossible to actively do so: Any institution that does so would do so at its own expense. Universities look out to their own interests, and their interest is to produce graduates. It then follows that no university would voluntarily do this. So long as there is a demand for the product of a PhD degree someone somewhere will supply it. Even if we force all domestic institutions to limit the number of graduates, the market will still be flooded with foreign-trained graduates.
The poor job market is reducing the value of a PhD all on its own, and degree-holders will reduce accordingly.

3) Wage controls for post-docs, though also probably impossible for the reason above, will do what wage controls do everywhere else: Increase unemployment. The problem of excess post-docs follows, unsurprisingly, from the problem of too many PhD students.

4) The article alludes to the problem of poor reproducibility in medical research, and correctly asserts that it follows from a hyper-competitive environment which incentivizes sloppy work, and dis-incentivizes fact-checking. For those interested in the actual number, there are several studies out there (notably by Amgen and Bayer) which show that landmark biomedical research is only about 20% replicable.

5) Academia produces a staggering volume of misinformation. The success of Phase II clinical trials, wherein the efficacy of the compound is tested, has fallen to less than 20%. Drug development costs are paid by the end-user. The cost of poor medical research is therefore also showing up indirectly in our runaway health-care costs which, contrary to the popular myth, are primarily not demographically driven.

The OECD spends about 60b$/yr on basic research. When the taxpayer learns what poor value he's getting from his expenditure there will be a strong reaction, and probably an over-reaction. There is a problem with a flooded labour market now, but it will probably get much much worse

Permalink to Comment

24. Anon on April 17, 2014 11:45 AM writes...

It looks like my previous note did not go through because of the link. Below is the NIH's response to this issue after a committee raised alarms several years in a row.

"NIH has undertaken the following initiatives to make progress towards achieving these goals between now and 2015:

1.Establish a grant program to encourage innovative training approaches
2.Improve graduate student and postdoctoral researcher training
3.Develop a simple and comprehensive tracking systems for trainees
4.Encourage fair consideration of staff scientists on grant applications
5.Initiate discussion with the community to assess NIH support of faculty
6.Create an office in the NIH Office of the Director to assess the biomedical research workforce
7.Conduct ACD Working Group study on optimal research training of individuals in clinical disciplines"

To real changes, basically just your general business speak and the inaction you might expect.

Permalink to Comment

25. Anonymous on April 17, 2014 11:46 AM writes...

I've spent lots of time thinking about this. I don't really have a solution other then noticing that the system we put our professors in is flawed. They will do what's best for them.

Germany is a nice example to emulate, for a strong chemical industry.

I still think we need the fundamental research to push advancements in science, but I would like to see, if we could somehow lower the barrier to entry for chemical start ups.

Institutes or in